Aim 1. While the observation that sleep is conserved suggests that it provides important fitness benefits, the relationship between sleep and evolutionary fitness is not known. We have developed a novel approach to investigate the possible effect of sleep on reproductive fitness by quantifying the phenotypic and genetic co-variation between sleep and traits that are important determinants of reproductive success in Drosophila melanogaster. Different morphological, behavioral, and life-history factors may contribute to reproductive success. The first trait we are focusing on is the ovarioles, tubes in the fly ovary that produce oocytes. Numbers of ovarioles strongly affect reproductive fitness in female flies by limiting maximum fecundity. We counted the numbers of ovarioles in females of the Drosophila Genetic Reference Panel (DGRP), a collection of inbred lines derived from a wild North American population. Ovariole number varies significantly among lines in the DGRP and has a strong heritable component. We identified polymorphisms significantly associated with ovariole number that could be localized to 56 putative candidate genes. While none of the ovariole number polymorphisms overlapped with sleep-associated polymorphisms, overlap occurred at the gene level. We identified 19 common genes having polymorphisms associated with both ovariole number and sleep. We tested mutations in these 19 genes for their effects on ovariole number and sleep; we are tested mutations in the remaining 37 genes for their impact on ovariole number. We hypothesize that if a mutation in a candidate gene significantly alters ovariole number and/or sleep, the SNP originally associated with the gene may alter the gene's function to affect these traits. We found mutations that conferred both quantitative and qualitative effects on ovariole number. For two mutants that we examined, severe morphological defects in ovariole number were observed, though the morphological defect was not completely penetrant in one of the mutants. We measured the reproductive success of these two mutations. We also demonstrated that mutations in overlapping candidate genes from the genome-wide association can affect both ovariole number and sleep. Aim 2. Our genome-wide association study (GWAS) of sleep in flies indicated that there were several candidate genes with homology to genes previously implicated in human sleep studies. We tested nine of these genes using RNAi-mediated knockdown. Six of these genes have known roles in fly neural development and are expressed at different life stages of the fly. In order to determine whether gene expression at different developmental stages was important for adult sleep, we used a conditional GeneSwitch driver to reduce the function of these candidate genes in neurons at different stages of development. Preliminary data indicates many significant effects on sleep that are highly sex-specific and dependent upon the developmental stage of the knockdown. We will complete these experiments and verify that the knockdown protocol was effective. Aim 3. Our previous GWAS of sleep in Drosophila revealed 8,127 polymorphisms associated with fourteen sleep traits. Some of these polymorphisms fall within genes, and some are intergenic. For those polymorphisms falling within genes, we have identified those with the greatest homology to human genes. We will further explore the role of these genes in sleep using flies. We are currently testing mutations in 10 genes that were identified in our Drosophila GWA study as well as a study of sleep in humans. Promising candidate genes from these fly studies will eventually be tested in a mouse model. Sleep-associated polymorphisms falling into intergenic regions are more challenging, as the sequence between genes is not expected to be conserved across species. We have therefore undertaken a preliminary in silico approach to determine if any of these polymorphisms impact conserved aspects of gene regulation. We queried the Drosophila Encyclopedia of DNA Elements (ModENCODE) data base for regulatory modifications that map to our genome-wide association data. We found a paucity of night sleep-associated polymorphisms in 4 types of histone modification sites thought to be activating, and enrichment of waking activity-associated polymorphisms in sites associated with the same histone modifications. We also found enrichment of night sleep-associated polymorphisms in one type of histone modification site thought to be repressive. We hypothesize that these sites regulate gene expression. To test this hypothesis, we are currently assessing transcript abundance in genes proximal to these sites.